Four serotypes of dengue virus (DENV-1-4) that circulate globally, causing more human illness than any other arthropod-borne virus. DENV infection results in dengue fever, an acute febrile illness, or the more severe, life- threatening dengue hemorrhagic fever/dengue shock syndrome. Previous work had proposed that host immune responses to sequential infections by different DENV serotypes may exacerbate disease. More recent work suggests that other factors, such as the genetics of the infecting viral strain, also play a major role in determining disease severity as well as viral fitness. Viral genetic differences can affect the inherent virulence of the virus, its susceptibility to neutralization by pre-existing anti-DENV antibodies in the human host, and/or viral fitness in its mosquito vector. In the larger context of DENV evolutionary dynamics, the relative abundance of different serotypes and viral lineages is continually changing in the face of host immunity, reflected in the local extinction and emergence of viral clades (i.e., 'clade replacement'),and due to introduction of new viruses from different regions. The central aim of our project is therefore to reveal the evolutionary processes that act on DENV within both endemic and epidemic populations and in a changing immunological landscape. We will use a unique combination of comparative and experimental techniques, combined with an unprecedented level of complete genomic sequence data, to determine whether viral lineages differ in fitness and, importantly, to reveal the determinants of fitness. A key intellectual advance of the work proposed here is to determine whether viral lineages co-circulating within a single population differ in fitness and if this fitness is dependent on the other co-circulating serotypes. Such analysis is of fundamental importance because of forthcoming dengue vaccination. Our overall hypothesis is that the on-going evolution of DENV genomes gives rise to viral strains with greater fitness and virulence in vivo. We propose that the molecular mechanisms through which these genetic changes mediate their effects can be first explored using comparative genomics and then defined experimentally in relevant in vitro assays, established mouse models, and Aedes aegypti mosquitoes. In Specific Aim 1, we will determine the evolutionary dynamics of DENV and identify all examples of clade replacement in both endemic (Viet Nam) and emergent (Nicaragua) populations. In Specific Aim 2, the in vitro replication dynamics of distinct DENV lineages from Nicaragua and Viet Nam will be characterized and the determinants of increased virulence and/or fitness will be identified. In Specific Aim 3, we will evaluate the susceptibility of the distinct DENV clades to neutralization by pre-existing heterotypic immunity in vitro and in a mouse model. Finally, fitness differences between distinct DENV lineages in A. aegypti mosquitoes will be evaluated in Specific Aim 4. This work has important implications for dengue biology, evolution, and vaccine development, as well as for the evolutionary dynamics and emergence of antigenically variable pathogens in general. PUBLIC HEALTH RELEVANCE: We have a unique opportunity to combine sequence analysis of numerous full-length dengue virus genomes with experimental approaches in vitro, in a mouse model, and in mosquitoes to identify the underlying mechanism of increased viral virulence and fitness in the context of pre-existing host immunity and to identify the genetic determinants responsible. Ultimately, an understanding of the molecular and evolutionary bases of viral fitness is essential for developing and implementing vaccines and drugs to control dengue. More generally, this work this will provide unique insights into the dynamics and determinants of viral evolution and emergence.